Is dabigatran an appropriate periprocedural anticoagulant? Review of recent data and HRS 2012 abstracts

May 17, 2012 06:52 EDT

Catheter ablation of atrial fibrillation has come a long way.

Although success rates remain modest, the safety of the procedure has advanced considerably. Stroke rates, in particular, have decreased. Reasons for this are speculative but likely include widespread adoption of saline-irrigated ablation catheters, improved periprocedure anticoagulation, and progressing skill of operators.

Many years ago, my institution adopted a policy of doing AF ablation during uninterrupted warfarin. This allowed us to avoid transesophageal echocardiograms (in most patients) and reduced our stroke rate to near zero.

Now we have a problem. What should be done with the AF patient doing well on dabigatran (and soon rivaroxaban) who is referred for AF ablation? Catheter ablation of AF is not a cardioversion. Placing sheathes in the left atrium and burning atrial endothelium creates an especially thrombotic milieu. The safety of AF ablation depends greatly on the adequacy of anticoagulation before, after, and during the procedure.

Increasing numbers of AF patients are treated with the new anticoagulants. Despite media reports to the contrary, many AF patients are doing well on the new drugs—just like the tens of thousands in the clinical trials.

But then they come for ablation—for burns and sheaths in the left atrium. What do we do now? Do we insist on warfarin, or can we leave them on dabigatran?

Pre-HRS data on periprocedure dabigatran

After this multicenter trial came out strongly negative against periprocedure dabigatran, some have urged that all patients considered for ablation be treated with warfarin. A word on this highly influential trial: In eight centers, 145 patients underwent AF ablation with dabigatran as the anticoagulant. They were compared with a matched but not randomized group of patients on uninterrupted warfarin. No patient on warfarin had a stroke or transient ischemic attack (TIA), while three patients on dabigatran had an embolic event (p=NS). Major bleeding occurred in 6% of the dabigatran group vs 1% with warfarin (p=0.019). Using a composite end point of embolic event plus stroke, the researchers concluded dabigatran was inferior to warfarin in the periprocedure period.

Others have reported less negative data on dabigatran. At the AHA meeting in November of 2011, a group from Cleveland Clinic presented a series of 101 patients who had AF ablation on either dabigatran or warfarin. They reported no significant embolic events or bleeding in either group.  

The practical problem, of course, is that switching patients' blood-thinning regimen in the real world doesn't always work as well as it does in centers blessed by foundation-supported armies of physician extenders. As shown in the ROCKET-AF trial, the risk of embolic events rises when anticoagulation is interrupted.

Not all AF centers have adopted a warfarin-only policy. That was obvious when I reviewed the many posters at HRS 2012 that studied dabigatran use during AF ablation.

I’d like to offer a quick review of the eight studies presented at HRS 2012 that considered periprocedural dabigatran during AF ablation. They were all small trials with slightly different protocols. This makes it tough to compare, but perhaps a picture might emerge. I’ve even added some of my own summary statistics at the end. (A Mandrola meta-analysis?)

Summary of dabigatran v warfarin (in AF ablation) at HRS 2012

First: Researchers from four major AF centers in the US compared a group of 178 patients on uninterrupted warfarin vs dabigatran. Strokes were reported in two patients in the dabigatran group and none in the warfarin arm. Major bleeds occurred in three patients in both groups—though one dabigatran bleed required transfusion. They concluded equivalence but worried about nonreversibility of dabigatran.

Second: A Japanese center reported a series of 503 patients separated into three groups—uninterrupted warfarin, warfarin with bridging heparin, and dabigatran. Strokes were not observed in any of the groups. More bleeds (3%) were seen in the bridging group. They concluded that dabigatran has an acceptably low bleeding risk during AF ablation.

Third: Vanderbilt researchers performed a retrospective review of 171 AF ablations—110 done with warfarin and 61 with dabigatran. Embolic complications occurred in two patients taking dabigatran and one on warfarin. Major bleeds were seen in only one patient in the dabigatran group, but five were noted with warfarin. Noteworthy here were that both dabigatran-related events occurred in patients who had delays in receiving the drug after the procedure. Also, three (of five) patients with subtherapeutic INRs who received heparin bridging suffered major bleeds. They concluded dabigatran compared favorably with warfarin and delay in giving dabigatran after the procedure may increase stroke risk.

Fourth: The Florida division of the Cleveland Clinic reported a consecutive series of 118 AF ablations—58 with warfarin and 60 with dabigatran. Embolic events were seen in none of the dabigatran-treated patients and in one patient on warfarin. Both groups had one patient each with a major bleed. Noteworthy was that intracardiac echo (ICE) identified two patients in the dabigatran group who had thrombus noted on a left atrial sheath. In both cases, the clot was aspirated out of the sheath. They concluded the strategies were equivalent but worried about the visual thrombus in the dabigatran group.

Fifth: University of South Carolina researchers retrospectively compared 282 AF ablations—169 with warfarin bridging and 113 with dabigatran. Embolic complications were seen in two patients in both groups. Major bleeds occurred in one patient in the warfarin group and none were seen with dabigatran. Minor bleeds, however, occurred in 19.5% of warfarin-treated patients and only 4% of those on dabigatran. They concluded dabigatran was the superior strategy. 

Sixth: This small study from the Florida branch of the Mayo Clinic compared only bleeding rates of warfarin and dabigatran during AF ablation—124 patients with warfarin and 31 with dabigatran. They reported no major bleeding and an equivalent number of minor bleeds in each group.

Seventh: Mount Sinai (NY) researchers report a series of 65 dabigatran-treated patients who underwent AF ablation (no comparison group). They report two patients suffered stroke or TIA (3%) and two had femoral hematomas. They called these findings significant and recommended further study.

Eighth: University of Alabama researchers report a series of 170 patients treated with uninterrupted dabigatran (11% of patients had one dose held the morning of procedure) during AF ablation. All patients had a transesophageal echocardiogram (TEE), and one patient was not ablated due to left atrial appendage (LAA) thrombus.  One patient had a TIA that resolved quickly without an abnormality on MRI, and two patients had nonlethal bleeding complications. They concluded that uninterrupted dabigatran was safe and effective.

I did a little "Mandrola" (unofficial) statistics on the six abstracts that compared dabigatran and warfarin and came up with this table.

Anticoagulant	Total patients presented, n	Stroke/TIA, n (%)	Major bleeds, n (%)
Warfarin total	1036	4 (0.3)	21 (2.0)
Warfarin unbridged	673	2 (0.3)	14 (2.1)
Dabigatran	549	6 (1.1)	6  (1.1)

Simple chi-square tests yielded a nonsignificant (p=0.08) increase in major bleeds with warfarin. Although surely not statistically pure, stroke/TIA events occurred more often in patients treated with dabigatran (p=0.04).

Dabigatran effects on heparinization

There's one more abstract on dabigatran that I’d like to tell you about. Researchers from Mid-America Heart Institute (Kansas City)  report that dabigatran-treated patients may require higher doses of heparin to achieve acceptable activated clotting times (ACT). When standard doses of heparin were used during AF ablation, dabigatran-treated patients showed longer times to acceptable ACT levels, and 33% of patients never reached target ACT. These observations are consistent with findings from other labs.

Conclusions

With only one published trial and a smattering of abstracts, it's hard to make strong statements about what anticoagulation strategy works best before and after AF ablation. Let me offer my best summary statements:

• The strategy of bridging with heparin in warfarin-treated patients looks to be the least desirable.
• Compared with warfarin, dabigatran is definitely not a superior periprocedure anticoagulant.
• The question of noninferiority of dabigatran awaits a randomized controlled trial. The pre-HRS data suggesting a signal for increased risk of embolic events with periprocedural dabigatran is less than robust.
• * A summation of periprocedural dabigatran trials at HRS 2012 confirms a very weak (if any) stroke/TIA signal with dabigatran. It’s definitely not clear.
• Major bleeding looks about the same with both agents.

 

In this ablationist's opinion, it is too soon to disqualify dabigatran as a periprocedure anticoagulant. One must weigh the possible small signal for increased stroke against the risks of switching anticoagulants right before a major procedure. Until randomized trials are done or more data become available, we simply don’t know the answer. Look for increasing use of rivaroxaban (and its many available doses) along with soon-to-be-approved apixaban to cloud an already cloudy picture.

JMM